Elsevier Editorial System(tm) for Colloids

and Surfaces A: Physicochemical and Engineering Aspects

Manuscript Draft

Manuscript Number: COLSUA-D-15-01762R1

Title: Thermodynamic and kinetic characterization of pH-dependent

interactions between bovine serum albumin and ibuprofen in 2D and 3D

systems

Article Type: Research Paper

Keywords: thermodynamics; kinetic studies; BSA; ibuprofen; nanocomposite;

SPR

Corresponding Author: Prof. Imre Dékány, DSc.

Corresponding Author's Institution: University of Szeged

First Author: Edit Csapó, PhD

Order of Authors: Edit Csapó, PhD; Ádám Juhász; Noémi Varga; Dániel

Sebők, PhD; Viktória Hornok, PhD; László Janovák, PhD; Imre Dékány, DSc.

Abstract: The interactions between bovine serum albumin (BSA) and

ibuprofen (IBU) were investigated at pH 3.0 and pH 7.4 by several two-

(2D) and three-(3D) dimensional techniques to provide quantitative,

kinetic and thermodynamic data on the BSA-IBU binding. Based on the

results, the preparation of BSA-IBU composite nanoparticles (NPs) were

successfully carried out for controlled drug release. The high resolution

transmission electron microscopy (HRTEM), dynamic light scattering (DLS)

and small angle x-ray scattering (SAXS) studies confirm the formation of

nearly monodisperse NPs with daverage = 10-13 nm depending on the protein

concentrations and IBU contents. The kinetics of pH-induced drug release

was studied by a vertical diffusion cell at pH 7.4 at 25 oC. The pH-

dependent changes in the secondary structure of BSA were proven by SAXS,

DLS and surface plasmon resonance (SPR) investigations. Depending on the

protein conformations, the SPR results suggest that the bonded amounts of

the drug molecule are 1239 mg IBU/g BSA and 174 mg IBU/g BSA at acidic

and neutral pH, respectively. Besides quantification of the interactions,

the rate of association (ka) and dissociation (kd), the KA and KD

standard equilibrium constants and the binding free energy (ΔG°) were

also calculated on the basic of SPR measurements. The ΔG° = -21.5 ± 0.2

kJ mol-1 obtained by SPR in 2D system is in good agreement with the ΔG° =

-17.38 ± 0.54 kJ mol-1 determined by isotherm titration calorimetry (ITC)

in solution (3D).

Prof. Dr. Veronique Schmitt

Special Issue Managing Guest Editor of 29th

ECIS 2015 (Bordeaux)

Colloids and Surfaces A: Physicochemical and Engineering Aspects

27 May, 2016.

Dear Prof. Schmitt!

According to some comments received from Editor and Reviewer#1 we have revised our

manuscript (COLSUA-D-15-01762, Thermodynamic and kinetic characterization of pH-

dependent interactions between bovine serum albumin and ibuprofen in 2D and 3D systems

Authors: E. Csapó*, Á. Juhász, N. Varga, D. Sebők, V. Hornok, L. Janovák, I. Dékány

*

Attached please find the responses to Editor and Reviewer suggestions and questions.

In the name of all co-authors I would like to thank you for the time and efforts while treating

our submission.

Yours sincerely,

Imre Dékány and Edit Csapó

corresponding authors

*Revision Letter/Notes

Response to each point of the comments of the Reviewer

We are very grateful to the Editor and Reviewer for their efforts to improve our

manuscript. Below, we give our point-by-point responses to the points raised by the

Reviewer, and also the changes made in the manuscript.

In response to the comments of the Reviewer#1.

1. „ The authors decided to study the influence of pH by using two pH values; one in acidic

region (pH = 3.0) and one in neutral pH region (pH = 7.4). However, somewhere in the

Introduction the reasons for choosing these two pH values should be stated and discussed.”

According to the request of Reviewer#1 the Introduction (page 2, line 27-32) were completed

with the followings:

„In the present work, BSA-IBU composite NPs were prepared at pH 3.0 for pH-induced

controlled drug release and kinetics of the ibuprofen release process at pH 7.4 was studied in

in vitro experiments. Since the preparation of composite NPs was carried out at pH 3.0 and

the drug release was measured at pH 7.4, the interactions between the protein and drug

molecule were investigated at the above mentioned two pH values by using several 2D and

3D techniques in order to provide deeper information on the binding and release processes.”

2. „ The equations (8) and (9) are basic thermodynamic equations. Maybe they can be

omitted form the text.”

We accept the comment of Reviewer#1 and the equations (8) and (9) are not presented as

equations, the formulas were inserted in the text (page 5, line 28-29).

3. „ The results obtained by calorimetric titrations are presented as DH = -22.85±0.57 kJ

mol-1 and 19.57±0.82 kJ mol-1. Were these calorimetric experiments repeated as in the

case of SPR measurements? If so, how many times?”

We accept the question of Reviewer#1. As it was mentioned in the 2.3, 2.4 and 2.6 sections

parallel measurements were carried out for DLS (3-times), SPR (2-times) and release kinetic

(2-times) as well. Naturally, for ITC experiments the titrations were repeated twice, and Fig.

5. shows only one representative calorimetric titration curves of IBU with BSA solution at pH

3.0 (a) and 7.4 (b) at 25 °C. The experimental errors were summarized in Table 1 and the text

was completed with the following sentence (page 5, line 1):

“Two parallel measurements were carried out.”

4. „In the text all thermodynamic state functions <DELTA>G, <DELTA>H and

<DELTA>S are not presented as standard quantities. However, in Table 1. the same

quantities are presented as standard quantities (<DELTA>G°, <DELTA>H° and

<DELTA>S°), while in the same Table the corresponding equilibrium constant KA is not

presented as standard. That should be checked.”

We accept the comments of Reviewer#1 and the text were corrected. Naturally, the

thermodynamic state functions are standard values as presented in Table 1. (corrections: page

1, abstract; page 3, line 1; page 8, line 27; page 10, line 18, 23, 25, 26). However, the KA is

also

standard value, but according to IUPAC nomenclature both the KA and the KA° are also

suitable. In this manuscript we used the KA.

5. „English should be improved: there are several misprints and grammatical errors. Here

are just few examples: Chapter 3.1"..sensorgrams is presented..." should be "… are

presented…" Chapter 3.3 "It has to mention…" should be corrected etc.”

According to the request of Reviewer#1 the text were corrected; numerous misprints and

grammatical errors were also corrected.

e.g. page 6, line 13 (“as” – “since”; page 6, line 21 (“do” – “does”); page 7, line 10 (“is” –

“are”); page 7, line 17 (“were” – “was”); page 8, line 16 (“it has to mention” – “because”);

page 8, line 25 (“The” – “A representative”); page 9, line 15 (“occurred” – “measured”); page

11, line 12 (“play” – “plays”); page 12, line 18 (“in” – “on”).

The originally submitted version of Graphical abstract, Highlights and Figures and Table were

not changed.

Thermodynamic and kinetic characterization of pH-dependent interactions

between bovine serum albumin and ibuprofen in 2D and 3D systems

E. Csapó1,*

, Á. Juhász1, N. Varga

2, D. Sebők

2, V. Hornok

2, L. Janovák

2, I. Dékány

1,*

Graphical abstract

(not proportional representation)

*Graphical Abstract (for review)

Thermodynamic and kinetic characterization of pH-dependent interactions

between bovine serum albumin and ibuprofen in 2D and 3D systems

E. Csapó1,*

, Á. Juhász1, N. Varga

2, D. Sebők

2, V. Hornok

2, L. Janovák

2, I. Dékány

1,*

Highlights

- Design of BSA-IBU NPs was carried out by the results of several 2D and 3D experiments

- The pH-induced structural changes of BSA were proven in 2D and 3D systems

- Quantitative data of the BSA-IBU interactions were presented at different pH

- Kinetic constants and thermodynamic state functions were determined by SPR and ITC

- The pH-induced ibuprofen release of the nanosized composite particles was confirmed

*Highlights (for review)

Thermodynamic and kinetic characterization of pH-dependent interactions

between bovine serum albumin and ibuprofen in 2D and 3D systems

E. Csapó1,*

, Á. Juhász1, N. Varga

2, D. Sebők

2, V. Hornok

2, L. Janovák

2, I. Dékány

1,*

1 MTA-SZTE Supramolecular and Nanostructured Materials Research Group, University of

Szeged, Department of Medical Chemistry, Faculty of Medicine, H-6720 Dóm tér 8, Szeged,

Hungary

2 Department of Physical Chemistry and Material Sciences, University of Szeged, H-6720

Aradi Vt. tere 1, Szeged, Hungary

*Corresponding authors at: MTA-SZTE Supramolecular and Nanostructured Materials Research

Group, University of Szeged, Hungary, E-mail addresses: tidecs2000@yahoo.co.uk (E. Csapó),

i.dekany@chem. u-szeged.hu (I. Dékány) Tel: +36(62)544210, Fax: +36(62)544042

Abstract

The interactions between bovine serum albumin (BSA) and ibuprofen (IBU) were

investigated at pH 3.0 and pH 7.4 by several two-(2D) and three-(3D) dimensional techniques

to provide quantitative, kinetic and thermodynamic data on the BSA-IBU binding. Based on

the results, the preparation of BSA-IBU composite nanoparticles (NPs) were successfully

carried out for controlled drug release. The high resolution transmission electron microscopy

(HRTEM), dynamic light scattering (DLS) and small angle x-ray scattering (SAXS) studies

confirm the formation of nearly monodisperse NPs with daverage = 10-13 nm depending on the

protein concentrations and IBU contents. The kinetics of pH-induced drug release was studied

by a vertical diffusion cell at pH 7.4 at 25 oC. The pH-dependent changes in the secondary

structure of BSA were proven by SAXS, DLS and surface plasmon resonance (SPR)

investigations. Depending on the protein conformations, the SPR results suggest that the

bonded amounts of the drug molecule are 1239 mg IBU/g BSA and 174 mg IBU/g BSA at

acidic and neutral pH, respectively. Besides quantification of the interactions, the rate of

association (ka) and dissociation (kd), the KA and KD standard equilibrium constants and the

binding free energy (ΔG°) were also calculated on the basic of SPR measurements. The ΔG° =

-21.5 ± 0.2 kJ mol-1

obtained by SPR in 2D system is in good agreement with the ΔG° = -

17.38 ± 0.54 kJ mol-1

determined by isotherm titration calorimetry (ITC) in solution (3D).

*ManuscriptClick here to view linked References

Keywords: thermodynamics, kinetic studies, BSA, ibuprofen, nanocomposite

1. Introduction

Nanoscale drug delivery systems have been under investigations for several decades [1-3]. At

present, numerous types of NPs are designed as feasible candidates for gene therapy and

molecular imaging [4], but only very few have actually to mature to clinical applications.

Polymer-, dendrimer-, lipid-, iron oxide-, quantum dots- or other organic- and inorganic-based

NPs are synthesized in order to deliver a drug to the right place at the right time in adequate

concentration [5-8]. The proteins are also widely used for encapsulation and transportation of

different drug molecules. The albumin-(BSA or HSA)-based NPs play a determinant role in

the development of novel nanocarrier systems because many binding sites are available to

several drug molecules. Moreover, the albumins have various specific advantages in nano-

scale range, such as biodegradability, biocompatibility and non-toxicity [8]. In the interest of

the development of an effective drug delivery systems the interaction between the drug and

the carrier should be strong enough to facilitate the transport but also weak enough to release

the drug to the target. Thus, the quantitative study of the binding thermodynamics and the

knowledge of the kinetics of the release process are necessary [9]. Contrary to the (radio)-

labelled techniques in recent years a number of “label-free” techniques have been developed

to report biomolecular interactions [10-12]. Two-dimensional SPR is a label-free technique

and capable of measuring real-time quantitative binding affinities and kinetics for proteins

interacting with biomolecules using relatively small (in nanomolar range) quantities of

materials and has potential to be medium-throughput [13-15]. The conventional SPR

technique requires that one binding component to be immobilised on a sensor chip while the

other binding component in solution is flowed over the sensor surface; a binding interaction is

detected using an optical method that measures remarkably small changes in refractive index

at the sensor surface. By using this biosensor assay not only quantitative and kinetic

information can be obtained, but the thermodynamic state functions of the interactions as well

[16,17] because the experiments are carried out at different temperatures.

In the present work, BSA-IBU composite NPs were prepared at pH 3.0 for pH-induced

controlled drug release and kinetics of the ibuprofen release process at pH 7.4 was studied in

in vitro experiments. Since the preparation of composite NPs was carried out at pH 3.0 and

the drug release was measured at pH 7.4 the interactions between the protein and drug

molecule were investigated at the above mentioned two pH values by using several 2D and

3D techniques in order to provide deeper information on the binding and release processes.

SPR and SAXS measurements were carried out to study the size and the structure of the

nanosized particles and to determine the binding capability of protein at different pH. The

thermodynamic binding constant (Kb), the state functions (ΔG°, ΔH°, ΔS°) and also the

stoichiometry of the interaction (n) were determined by ITC [18], while the rate of association

and dissociation and the KA and KD standard equilibrium constants were calculated by fitting

of the SPR sensorgrams. The calculated kinetic constants obtained by SPR in 2D systems

were compared with the results of the IBU release process measured in aqueous solution (3D).

2. Materials and Methods

2.1 Materials

All chemicals and solvents were of analytical grade and were used without further

purification. The BSA (fraction V), the IBU (C13H18O2) and the components of the

McIlvaine’s buffer (pH 3.0) and the phosphate buffer (PBS, pH 7.4) were purchased from

Sigma Aldrich, the sodium chloride (NaCl), the sodium sulphate (Na2SO4), the sodium

hydroxide (NaOH) and the hydrogen chloride (HCl) from Molar Chemicals. The stock

solutions were freshly prepared, using Milli-Q ultrapure water (18.2 MΩ cm at 25 °C).

2.2 Preparation of BSA-IBU nanocomposite particles

The studied nanosized protein-non steroidal anti-inflammatory (NSAID) composites were

prepared according to the procedure published previously [19,20]. Briefly, 20 w/v% BSA was

dissolved in 15 ml buffer solution (McIlvaine buffer, pH 3.0). When it completely dissolved

IBU molecules were added to the BSA solution with continuous stirring in 1:1 and 1:10 molar

ratios. We have stirred the solution for two additional hours at room temperature in order to

form the BSA-IBU nanocomposites since more and more drug binding results more and more

dissolved drug molecules. The BSA-IBU NPs were precipitated by 2M Na2SO4. The product

was obtained by freeze drying (lyophilized) after centrifugation (15000 rpm, 15 min).

2.3 HRTEM, DLS and SAXS measurements

HRTEM images were taken by a FEI Tecnai G2 20 X-TWIN microscope with tungsten

cathode at 200 kV. The parallel DLS measurements were performed with a Horiba,

Nanopartica SZ-100 Nanoparticle Analyzer (He-Ne laser with 532 nm wavelength) in order to

determine the size of NPs. Small angle X-ray scattering (SAXS) were used to analyze the

morphology, size and inner structure of the prepared composites and also to study the

conformation change of BSA at acidic and neutral pH values. SAXS curves were recorded

with a slit-collimated Kratky compact small-angle system (KCEC/3 Anton-Paar KG,

Graz, Austria) equipped with a position-sensitive detector (PSD 50M from M.Braun

AG. Munich, Germany). Cu Kα radiation was generated by a Philips PW1830 X-ray

generator operating at 40 kV and 30 mA.

2.4. SPR investigations

SPR measurements were carried out to determine the size and the orientation of BSA

adsorbed on the gold chip, and the binding capability of IBU on the BSA-functionalized gold

surface at pH 3.0 and pH 7.4. A two-channel SPR sensor platform developed at the Institute

of Photonics and Electronics (Prague) was used. The SPR chip is a thin gold layer (50 nm

thick) deposited on a glass substrate. During investigations, a flow rate of 25 µl·min− 1

was

applied at constant temperature (+20±0.1 oC). The interaction of IBU with the BSA was

studied in the concentration range 5.0-40.0 μM in PBS solution at pH 3.0 and 7.4 under

physiological conditions (150 mM NaCl). Parallel measurements were performed and the

standard deviations of the sorption experiments were ± 4.5%. In each step, 500 μl BSA (c =

50 μM) and 500 μl IBU solutions (c = 5.0; 10.0; 20.0; 30.0; 40.0 μM) were injected and the

sorption process (~ 20 min) was followed by rinsing with buffer. The SPR sensorgrams were

analyzed in real time by a special software package that allows determination of the resonant

wavelength in both sensing channels. Based on the results of 2D SPR experiments, the rate of

association and dissociation, the corresponding equilibrium constants and certain

thermodynamic state functions were determined according to the following evaluation

process. If the small molecules bind to the immobilised proteins, there is an association phase

during which binding sites become occupied and the positive slope of SPR curve can be used

to measure the rate of association (ka). When steady-state is achieved the RIU (refractive

index unit) value corresponds to the changed final critical angle (angle modulated type) or the

final wavelength of maximal plasmonic loss (wavelength modulated type). This maximum

RIU value relates to the concentrations of immobilised protein and analyte molecules and so

can be used to measure the binding affinity constant (KD = kd/ka). When small molecules are

removed from the continuous flow there is a dissociation phase during which binding sites

become unoccupied and the (negative slope) of curve can be used to measure the rate of

dissociation (kd).

2.5. ITC studies

Thermometric titrations were also performed using a MicroCal VP-ITC (Isothermal Titration

Calorimeter, MicroCal, USA) power compensation microcalorimeter with a cell volume of

1.4163 ml. The solutions were previously degassed by means of a vacuum degasser

Thermovac (MicroCal, USA). Two parallel measurements were carried out. The enthalpy

changes were recorded upon stepwise additions of BSA into the reaction cell containing

ibuprofen from a 300 μL syringe. Aliquots of 10 ml were injected at periodic time intervals

(10 s per injection, 5 min between injections). Blank experiments were performed in order to

make corrections for the enthalpy changes corresponding to the dilution of titrant. The

enthalpograms (calorimeter power signal, P(t) = dQ/dt vs. t) were evaluated with Origin® 7

software supplied by MicroCal. The area below each calorimeter peak i yielded a single point

in S-shaped reaction enthalpy curve [21]:

(1)

Analysis of the S-shaped curves provided the stoichiometry (n) of the reaction, the binding

constant (K), and the standard enthalpy of binding (H) according to the followings:

For a ligand X binding to a single set of n identical sites on a macromolecule M:

(2)

(3)

(4)

The binding constant (K) is:

(5)

(6)

where is the fraction of sites occupied by ligand X, Xt and [X] are bulk and free

concentration of ligand, n is the number of sites. The total heat content (Q) of the solution

contained at fractional saturation is:

(7)

where ΔH is the molar heat of ligand binding and Mt is the bulk concentration of

macromolecule in V0 active cell volume. The one set of sites model was applied for the curve

fitting after removing the first titration points and substracting the reference water curve. The

Gibbs free energy and the entropy term of the reaction was calculated via the well-known

basic thermodynamic equations (ΔG° = -RT lnK and ΔG° = ΔH°-TΔS°).

2.6. Release of IBU from BSA-IBU NPs

The obtained BSA-IBU composites were dispersed in PBS solution. 1.5 ml suspension was

filled to the vertical diffusion cell (Franz cell; HANSON CO.) above a cellulose membrane

(Sigma-Aldrich). Owing to the pH-changes, the IBU starts to diffuse through the membrane to

the pure PBS buffer. The concentration of the IBU was recorded by UV-1800

spectrophotometer at 264 nm. The absorbance spectra were taken every 10 min in the first

hour than once an hour to 500 min. Every measurement was repeated twice.

3. Results and discussion

3.1 The pH-dependent structure of BSA determined by DLS, SAXS and SPR experiments

It is well-known that the pH has a prominent role in the change of protein structure and

the substrate binding as well [22]. The knowledge of the exact structure of the protein at

different pH is necessary to design protein-based NPs for encapsulation of drug molecules.

Since the BSA-IBU NPs were fabricated at pH 3.0 and the spontaneous release of IBU from

NPs was measured at pH 7.4, the size and the structure of BSA at the previously mentioned

pH values were studied by DLS, SAXS and the two-dimensional SPR experiments. Figure 1

represents the Kratky (a), Guinier (b) plots and pair distance distribution functions (c) of BSA

colloid solutions at different pH. The Kratky plot (Ih2 vs. h, where I is the scattering intensity

and h is the scattering vector) representation provides information about the secondary

structure of the protein, it is informative about both the globularity and the flexibility of the

protein. In the case of folded globular protein, the Kratky plot will show a peak at low q

values. If the curve does not converge to the q-axis at high q that indicates that the protein has

a definite flexibility. Both the peak and the divergence from the q-axis refers dominantly the

presence of a flexible folded state of the protein (Fig. 1a, pH 7.4, cBSA = 20.0 w/v %).

Decrease in the pH causes the change of the shape of the Kratky plot (decrease of the peak,

Fig. 1a, pH 3.0). This shape preferably indicates the formation of an unfolded state of the

BSA. The Guinier plot (lnI vs. h2) (Fig. 1.b) and the pair distance distribution function

(PDDF) (Fig. 1.c) are suitable to determine the size and morphology of the molecules. The

radius of gyration (Rg) of the molecule can be calculated by linear fitting of the initial range

(hRg < 1.3) of the scattering curve in lnI vs. h2 representation (Fig. 1b). As it can be seen the

smaller radius of gyration (Rg = 2.02 nm) is determined for the BSA solution at pH 7.4 while

the larger Rg value (Rg = 2.64 nm) belongs to the 20.0 w/v % BSA solution at pH 3.0. These

data are in good agreement with the results of both Kratky and the PDDA representations as

well. On the whole, the detailed SAXS results strongly support that the BSA has an unfolded

(larger extension) structure at acidic pH (Fig. 1c, pH 3.0, the largest extension is 9.0 nm),

while the formation of a flexible folded structure (smaller extension) (Fig. 1c, pH 7.4, the

largest extension is only 6.3 nm) is confirmed at neutral pH. The DLS measurements also

confirm the above mentioned pH-dependent structural changes of the protein. Namely, the

measured average hydrodynamic diameter of BSA is d = 3.8±0.3 nm (polydispersity index,

PDI = 0.344) at pH 3.0, while d = 8.2±0.2 nm (PDI = 0.221) was obtained at pH 7.4. Besides

the classic three-dimensional techniques, two-dimensional SPR experiments were also

performed to get deeper information on the size and structure (orientation) of BSA at different

pH. The studied protein was immobilized onto the gold surface from aqueous solution (c =

0.05 mM) at 25 °C. The registered SPR sensorgrams are presented in Figure 2. As it can be

seen that 61.5 % and 70.0 % of adsorbed amount remains irreversibly bound at gold surface

after rinsing procedure at pH 3.0 and 7.4, respectively. Most probably the protein is bonded

onto the gold surface via cysteine residues resulting the formation of Au-S covalent bond. The

binding of BSA on gold surface caused ΔλpH 3.0 = 2.8 nm and ΔλpH 7.4 = 4.1 nm plasmon shifts

which correspond to the mspH 3.0 = 56.5 ng cm

-2 and

m

spH 7.4 = 86.0 ng cm

-2 adsorbed amount of

BSA. According to the calculation procedure published previously [15,23-25] the cross

sectional area (am/nm2) of the BSA at acidic and neutral pH values was determined. It was

found that the am for BSA is 195.4 nm2/protein

(pH 3.0) and

128.4 nm

2/protein

(pH 7.4) on

gold surface under the applied conditions. Taking into account that the protein covalently

binds onto the gold sensor surface the pH-dependent changes in the secondary structure of

BSA were proven in 2D systems as well indicating the larger cross sectional area of BSA is

obtained at pH 3.0. On the whole, the size of the BSA at acidic and at neutral pH determined

by SPR are in good agreement with the DLS and SAXS measurements.

3.2. Characterization of the size of the BSA-IBU NPs by HRTEM, DLS and SAXS results

Based on the results of SPR experiments, BSA-IBU composite particles were

successfully prepared at different BSA:IBU molar ratios. The Fig. 3.a shows the pair distance

distribution functions for pure BSA and BSA:IBU NPs at 1:1 and 1:10 ratios obtained by

SAXS. As it can be seen, the largest extension of the scattering objects continually increases

in the admixture of IBU to BSA solution. (pure BSA: dSAXS = 9.0 nm; BSA:IBU (1:1) NPs:

dSAXS = 10.0 nm; BSA:IBU (1:10) NPs: dSAXS = 10.5 nm). The sizes of NPs obtained by SAXS

are in good agreement with the HRTEM images and also the parallel DLS experiments. The

Fig. 3.c represents an image of BSA:IBU (1:10) NPs, the calculated average diameter is

dHRTEM= 12.9 ± 0.5 nm, while the parallel DLS curves (Fig. 3.b) indicate the formation of NPs

with dDLS= 11.8 ± 1.9 nm (average hydrodynamic diameter).

3.3 Quantification of the interaction between BSA and IBU by SPR experiments

The interaction of IBU with BSA were investigated by SPR to provide quantitative data

of the protein-biomolecule bindings. In order to determine the binding capacity of IBU on

BSA-covered gold surface the sorption/binding of IBU on protein-functionalized biosensor

chip was investigated at +25±0.1 °C in the concentration range of 5 – 40 µM. The registered

sensorgrams in Fig. 4. indicate that an increase in the concentration of the IBU solution from

5 µM to 40 µM results in a larger sorbed amount on the functionalized surface at both studied

pH (mspH 3 = max. 75-80 ng cm

-2, m

spH 7.4 = max. 12-15 ng cm

-2). Moreover, the experiments

clearly support that measurable high amount of IBU is bonded on BSA-covered gold surface

at acidic pH than that at pH 7.4. Namely, m = 1239 mg IBU is bonded to BSA (referring to

1.0 g serum albumin protein) at acidic pH while the bonded amount of IBU on BSA-covered

gold surface is only 0.174 mg at pH 7.4. Most probably the availability of hydrophobic

binding sites of BSA is more favoured for hydrophobic IBU molecules at acidic pH because

of the unfolded structure. Because the water solubility of IBU is relatively less at pH 3.0 and

above the concentration of 10 μM the aqueous solution is opalescent causing suddenly high

changes in the refractive index which is highlighted with grey colour in Fig. 4. Based on this

observation, only the effectively bound amount of IBU (signed with arrows) is used for the

calculations. The sensorgrams also confirm that the interaction between the IBU molecules

and the protein is fully reversible at pH 7.4 (Fig. 4. inset) because the adsorbed mass of IBU

fell to nearly zero on rinsing. The confirmation of this reversible interaction is crucial in order

to design potential nanocarrier composite system(s) for (controlled) drug release.

3.4 Thermodynamic characterization of BSA-IBU system by ITC

A representative calorimetric titration curves of IBU with BSA aqueous solution at pH 3.0

and 7.4 were reported in Fig. 5. It was found that the interaction between IBU and BSA in

aqueous solution both at pH 3.0 and 7.4 was exothermic process at 25 °C with H° = -22.85 ±

0.57 kJ mol-1

and -19.57 ± 0.82 kJ mol-1

, respectively. As expected, the value of n and K

indicated higher affinity at acidic pH with significantly more IBU molecules (values

summarized in Table 1) bound to BSA and the binding constant decreased from 3620 ± 89 M-

1 to 1110 ± 242 M

-1 with increasing pH. The relative magnitudes of the enthalpy and entropy

changes determine the resultant change in the Gibbs energy, which is thermodynamically

favoured, must be negative for a spontaneous process. The values of binding constants,

stoichiometry and the thermodynamic state functions are given in Table 1. The major driving

force or the interaction originates from the van der Waals interaction between the BSA and

the IBU molecules and the hydrophobic effect, the release of high energy water molecules

from the BSA cavity. The van der Waals interactions between the host and the guest species

are exothermic, accompanied by some entropy loss. In aqueous solution, apolar molecules are

surrounded by water molecules with a higher hydrogen-bond density relative to the hydrogen-

bond density of pure water. Upon interaction, molecules leave the solution and break this

hydration structure; the collapse of this “iceberg” structure is an endothermic process

accompanied by entropy production.

3.5 Kinetic studies in 2D and in 3D systems

Since the SPR signal responds directly to the amount of bound ligand in real time, it

provides a very powerful technique to study protein-ligand or any other biomolecular

interaction thermodynamics and kinetics. The protein-based nanocomposite containing IBU

was prepared at pH 3.0, but the release of the drug was measured at pH 7.4 under

physiological conditions. Based on it, the SPR sensorgrams registered at pH 7.4 was used for

detailed kinetic calculations. Assuming the A+B↔AB type reversible process is first-order

for each reactants in the association phase and the dissociation is also first-order thus the

following overall rate low can be given:

(8)

where [B] is the concentration of immobilized protein and [A] is the concentration of the

aqueous IBU solution, while the [AB] is the surface concentration of the BSA-IBU complex.

According to the research work of O’Shannessy et al. [26] integrated rate law in the

association process shows first order growth for the complex:

(9)

where the observed rate constant is:

(10)

In the dissociation phase, when only buffer is flowing over the BSA-covered gold surface,

([A]0≈0) the concentration of BSA-IBU complex decreases exponentially with a dissociation

rate constant:

(11)

A reasonable assumption is that the Δλ reading is proportional to the concentration of bound

complex, Δλ = α [AB]. Since the maximum concentration of the complex is [AB]max = [B]0,

we can determine the maximum Δλ value, Δλmax = α [AB] = α [B]0. Multiplying the Eq. (9)

and (11) by the proportional factor (α) the following equations for association and

dissociation reactions can be given:

(12)

(13)

Based on Eq. (12-13) and the registered sensorgrams, the rate constants were determined by

using nonlinear regression process. The Fig. 6.a shows the measured (black) and the fitted

(grey) sensorgrams. Rate constant of association and dissociation process can be determined

from the slope and the intercept of the observed rate constant vs. concentration of the aqueous

IBU solution plot (not presented here) according to Eq. (10). Linear regression analysis of the

previously mentioned kobs vs. cIBU plot results the following parameters: ka = 128 ± 9 M-1

s-1

, kd

= 0.02 ± 0.04 s-1

, KA = 5965 ± 420 M

-1 and ΔG° = -21.5 ± 0.2 kJmol

-1. As it can be seen, if we

use linear regression the intercept (kd) have high standard deviation. To avoid this uncertainty

the rate constant of dissociation was extracted from exponential curve fitting to Eq. (13) (Fig.

6.a). In this case the dissociation rate constant obtained by the nonlinear regression is kd =

6.1×10-3

± 3.7×10-4

s-1

. Despite of the above mentioned uncertainty of kd, the calculated value

for the Gibbs free energy (ΔG°) obtained by linear regression analysis shows good agreement

with the ITC experiments (Table 1). Namely, the Gibbs free energy of BSA-IBU complex

formation at pH 7.4 is ΔG° = -21.5 ± 0.2 kJmol-1

based on the observed rate constant using

kinetic analysis while ΔG° = -17.38 ± 0.54 kJmol-1

is determined according to ITC result.1

1 The Gibbs free energy of BSA-IBU complex formation originates from the difference between the

Gibbs free energy of the unbounded state (before binding) and the saturated state (after binding).

The release of the IBU from the composite NPs was measured in aqueous solution by using

vertical diffusion cell at 25 oC. The registered release profile was shown in Fig. 6.b. As it can

be seen that the release process starts slower in the case of the 3.9 w/v % sample [19] but after

500 min the dissolved amount of the IBU became equal. Various kinetic models (presented in

Table 2) were applied to describe the release mechanism of the IBU from NPs. The

correlation coefficient (R2) and the release rate (kd) values at 25

oC are summarized in

Table 2. The R2 values indicate the accuracy of the applied kinetic model. Although the

release profiles run same way, the release mechanism is definitely different. If we use BSA in

lower concentration - the protein-drug molar ratio is same in both cases - the release of the

drug molecule is independent on its concentration. If the BSA concentration is higher the

release mechanism follows very well the first-order rate model, which means that the drug

release rate depends on its concentration. The results show that the release mechanism of the

drug can be influenced by the BSA concentration, not only by forming shells around the

protein core.

4. Conclusion

The design of novel nanocarrier composite systems plays an important role in different

pharmaceutical developments. In order to produce potent nanosized composite particles

including drug molecules the holistic characterization of the interactions between the carrier

material(s) (e.g. proteins) and the drug agent(s) are crucial. Besides quantification of the

interaction, the quantitative study of the binding thermodynamics and the knowledge of the

kinetics of the drug release process are necessary. However numerous research group focus

their studies on the characterization of biomolecular interactions by SPR or ITC

measurements, we successfully used the two-dimensional SPR technique for kinetic and

thermodynamic evaluations draw a parallel between the results of individual measuring

techniques and theories using in 3D systems. We confirmed that the important physical-

chemical parameters of a protein-drug molecule interactions obtained by SPR are in good

agreement with the results of the classic three-dimensional SAXS, DLS, ITC and release

studies in solution. Based on the experimental results (quantitative, kinetic and

thermodynamic data) of this sensor technique our work may contribute to the development of

novel nanosized drug-carrier NPs.

Acknowledgement

The authors are very thankful for the financial support from the Hungarian Scientific

Research Fund (OTKA) K 116323.

References

[1] L. Brannon-Peppas, J. O. Blanchette, Nanoparticle and targeted systems for cancer

therapy, Advanced Drug Delivery Reviews, 64 (2012) 206–212.

[2] A. Kumari, S. Kumar Yadav, S. C. Yadav, Biodegradable polymeric nanoparticles based

drug delivery systems, Coll. Surf. B. 75 (2010) 1–18.

[3] C. Shi, D. Guo, K. Xiao, X. Wang, L. Wang, J. Luo, A drug-specific nanocarrier design

for efficient anticancer therapy, Nature Communications, 6 (2015) 7449.

[4] G. Liu, M. Swierczewska, S. Lee, X. Chen, Functional nanoparticles for molecular

imaging guided gene delivery, Nano Today 5 (2010) 524-539.

[5] J. Hannah Lee, Y. Yeo, Controlled drug release from pharmaceutical nanocarriers,

Chemical Engineering Science 125 (2015) 75–84.

[6] N. Sanvicens, M. P. Marco, Multifunctional nanoparticles – properties and prospects for

their use in human medicine, Trends in Biotechnology 26 (2008) 425-433.

[7] C. Deng, Y. Jiang, R. Cheng, F. Meng, Z. Zhong, Biodegradable polymeric micelles for

targeted and controlled anticancer drug delivery: Promises, progress and prospects, Nano

Today 7 (2012) 467-480.

[8] P.-L. Lam, S. H.-L. Kok, R. Gambari, T.-W. Kok, H.-Y. Leung, K.-L. Choi, C.-S. Wong,

D. K.-P. Hau, W.-Y. Wong, K. H. Lam, Z.-X. Bian, K. K.-H. Lee and C.-H. Chui, Evaluation

of berberine/bovine serum albumin nanoparticles for liver fibrosis therapy, Green Chem. 17

(2015) 1640-1646.

[9] J. Siepmann, F. Siepmann, Mathematical modeling of drug delivery, International Journal

of Pharmaceutics 364 (2008) 328–343.

[10] L.A.A. de Jong, D.R.A. Uges, J.P. Franke, R. Bischoff, Receptor-ligand binding

assays: Technologies and applications, J Chromatography B, 829 (2005) 1-25.

[11] R.J Ward, G. Milligan, Structural and biophysical characterisation of G protein-coupled

receptor ligand binding using resonance energy transfer and fluorescent labelling techniques,

Biochim Biophys Acta. 1838 (2014) 3-14.

[12] T. Heyduk, Measuring protein conformational changes by FRET/LRET, Curr. Opin.

Biotechnol 13 (2002) 292-296.

[13] H. Sípová, J. Homola, Surface plasmon resonance sensing of nucleic acids: a review,

Anal. Chim. Acta, 773 (2013) 9-23.

[14] W.A. Murray, W.L. Barnes, Plasmonic Materials, Adv. Materials, 19 (2007) 3771-3782.

[15] D. Sebők, E. Csapó, T. Preocanin, G. Bohus, N. Kallay, I. Dékány, Adsorption of

ibuprofen and dopamine on functionalized gold using surface plasmon resonance

spectroscopy at solid-liquid interface, Croat. Chem. Acta 86 (2013) 287-295.

[16] R.B M Schasfoort, A. J Tudos, Handbook of Surface Plasmon Resonance, Cambridge,

UK, 2008.

[17] D.G Myszka, Kinetic analysis of macromolecular interactions using surface plasmon

resonance biosensors, Current Opinion in Biotechnology, 8 (1997) 50-57.

[18] C. Ràfols, S. Zarza, E. Bosch: Molecular interactions between some non-steroidal anti-

inflammatory drugs (NSAID’s) and bovine (BSA) or human (HAS) serum albumin estimated

by means of isothermal titration (ITC) calorimetry and frontal analysis capillary

electrophoresis (FA/CE). Talanta 130 (2014) 241-250.

[19] N. Varga, M. Benkő, D. Sebők, I. Dékány, BSA/polyelectrolyte core–shell nanoparticles

for controlled release of encapsulated ibuprofen, Coll. Surf. B, 123 (2014) 616–622.

[20] M. Benkő, N. Varga, D. Sebők, G. Bohus, Á. Juhász, I. Dékány,

Bovine serum albumin-sodium alkyl sulfates bioconjugates as drug delivery systems, Coll.

Surf. B, 130 (2015) 126–132.

[21] M. Benkő, Z. Király, Thermodynamics of inclusion complex formation of b-cyclodextrin

with a variety of surfactants differing in the nature of headgroup. J. Chem. Thermodynamics

54 (2012) 211-216.

[22] R. Buratto, D. Mammoli, E. Chiarparin, G. Williams, G. Bodenhausen, Exploring weak

ligand-protein interactions by long-lived NMR states: improved contrast in fragment-based

drug screening, Angew. Chem. Int. Ed. Engl. 53 (2014) 11376-80.

[23] E. Csapó, Z. Majláth, Á. Juhász, B. Roósz, A. Hetényi, J. Tajti, G.K. Tóth, L. Vécsei, I.

Dékány, Determination of binding capability and adsorption enthalpy between Human

Glutamate Receptor (GluR1) peptide fragments and kynurenic acid by surface plasmon

resonance experiments, Coll. Surf. B, 123 (2014) 924-929.

[24] E. Csapó, F. Bogár, Á. Juhász, D. Sebők, J. Szolomájer, Zs. Majláth, G.K. Tóth, L.

Vécsei, I. Dékány, Determination of binding capability and adsorption enthalpy between

Human Glutamate Receptor (GluR1) peptide fragments and kynurenic acid by surface

plasmon resonance experiments: Part 2, Interaction of GluR1270-300 with KYNA, Coll. Surf.

B, 133 (2015) 66-72.

[25] N. Varga, E. Csapó, Z. Majláth, I. Ilisz, I.A. Krizbai, I. Wilhelm, L. Knapp, J. Toldi, L.

Vécsei, I. Dékány, Targeting of the kynurenic acid across the blood-brain barrier by core-shell

nanoparticles, Eur. J. Pharm. Sci., 86 (2016) 67-74.

[26] D.J. O’Shannessy, M. Brigham-Burke, K.K. Soneson, P. Hensley and I. Brooks,

Determination of rate and equilibrium binding constants for macromolecular interactions

using surface plasmon resonance: use of nonlinear least squares analysis methods, Anal.

Biochem, 212 (1993) 457-468.

Figure captions

Figure 1. Kratky (a), Guinier (b) plots and pair distance distribution function (c) of aqueous

BSA solutions at pH 3.0 and 7.4. cBSA = 20 w/v %.

Figure 2. Representative SPR sensorgrams of the binding of BSA onto gold surface at

different pH (cBSA = 50 μM, I = 150 mM (NaCl) in PBS, T = 25 °C, flow rate of 25 µL min-1

).

Figure 3. PDDA representations of BSA, BSA:IBU 1:1 and 1:10 NPs (a), the parallel DLS

distribution functions (b) and a representative HRTEM image of BSA:IBU 1:10 NPs (c) at pH

3.0.

Figure 4. Representative SPR sensorgrams of the binding of IBU onto the BSA-

functionalized gold surface at pH 3.0 and pH 7.4 (inset) (I = 150 mM (NaCl) in PBS, T = 25

°C, flow rate of 25 µL min-1

).

Figure 5. Representative calorimetric titration curves of IBU with BSA solution at pH 3.0 (a)

and 7.4 (b) at 25 °C.

Figure 6. The measured (black) and the fitted (grey) sensorgrams to evaluate the kinetics of

BSA-IBU interaction at pH 7.4 under physiological conditions (a) and the release profile of

the IBU from the NPs at 25 °C at pH 7.4 (b).

Table 1. Thermodynamic state functions, stoichiometry and binding constants determined

from ITC and SPR measurements (T = 25 °C).

Table 2. The rate of dissociation (kd) for the IBU at 25 °C by SPR (2D) and by release studies

in solution (3D).

Table 1.

# the values were determined by ITC experiments according to Fig. 5. (mg IBU/g BSA).

* on BSA-covered gold surface (mg IBU/g BSA) using cIBU = 30 µM.

BSA-IBU

interaction

bonded

amount of

IBU

KA

(dm3 mol

-1)

n ΔH°

(kJ mol-1

)

ΔG°

(kJ mol-1

)

TΔS°

(kJ mol-1

)

ITC pH 3.0 45.9 mg# 3.62×10

3±89 22.1±0.1 -22.85±0.57 -20.31±0.06 -2.54±0.57

pH 7.4 40.5 mg# 1.11×10

3±242 10.3±0.5 -19.57±0.82 -17.38±0.54 -2.19±0.98

SPR pH 7.4 174 mg* 5.97×10

3±420 n.i. n.i. -21.5±0.2 n.i.

Table 2.

IBU release kinetic models 3.9 % BSA-IBU 20.0 % BSA-IBU

Drug release kinetics

in solution (3D)

zero-order (s-1

) (independent from

concentration)

kd 6.48×10-2

±5.6×10-4

1.38×10-1

±5.2×10-3

R2 0.990 0.984

first-order (s-1

)

(dependent upon the

concentration)

kd 1.50×10-3

±9.9×10-4

2.40×10-3

±1.4×10-4

R2 0.989 0.995

Higuchi (s-1/2

) (diffusion-controlled)

kd 6.07×10-1

±1.2×10-1

3.65×100±7.8×10

-4

R2 0.964 0.992

SPR kinetic

experiments

(2D)

first-order (s-1

)

(dependent upon the

concentration)

kd 6.09×10-3

±3.65×10-4

R2 0.964

Figure 1.

Figure(s)

Figure 2.

Figure 3.

Figure 4.

Figure 5.

Figure 6.